NIHR Sheffield BRC
Equality Diversity and Inclusion 

Trial Forge Guidance 3: randomised trials and how to recruit and retain individuals from ethnic minority groups—practical guidance to support better practice

Randomised trials, especially those intended to directly inform clinical practice and policy, should be designed to reflect all those who could benefit from the intervention under test should it prove effective. This does not always happen. The UK National Institute for Health and Care Research (NIHR) INCLUDE project identified many groups in the UK that are under-served by trials, including ethnic minorities.

This guidance document presents four key recommendations for designing and running trials that include the ethnic groups needed by the trial. These are (1) ensure eligibility criteria and recruitment pathway do not limit participation in ways you do not intend, (2) ensure your trial materials are developed with inclusion in mind, (3) ensure staff are culturally competent and (4) build trusting partnerships with community organisations that work with ethnic minority groups. Each recommendation comes with best practice advice, public contributor testimonials, examples of the inclusion problem tackled by the recommendation, or strategies to mitigate the problem, as well as a collection of resources to support implementation of the recommendations.

We encourage trial teams to follow the recommendations and, where possible, evaluate the strategies they use to implement them. Finally, while our primary audience is those designing, running and reporting trials, we hope funders, grant reviewers and approvals agencies may also find our guidance useful.

Click here to access the full article. 

This text and examples were provided in January 2022 by Rosilda Panoni, In-Trans Coordinator at Grampian Regional Equality Council, UK (http://in-trans.co.uk).  

Here are some questions that may help a trial team think through their translation needs before speaking to a translation organisation: 

Identify your needs 

• How important is multilingual and communication for your project or organisation?

• Do you require face to face interpretation, translation of written text, or both? 

• How many languages do I need to translate into? 

• Try to get information about the language’s distribution in your target area and consider that alongside the focus of your project 

Translation project

• Which material should be translated first? 

• Which target languages do you want to prioritise? Is it critical to translate all content into all relevant languages at the beginning? 

• For example: You want to translate a questionnaire into Latvian, Lithuanian, Russian and Polish. Since Latvians and Lithuanians very often speak Russian, you might consider first translating the questionnaire into Russian and monitoring the requests for Latvian and Lithuanian, perhaps adding a line on the Russian translation about the possibility of translating the material into Latvian and Lithuanian. A good quality translation service will provide you this advice if you arrange a discussion in the planning stages of your project. 

• Another example will be, translate questionnaires first and check the responses and interaction of your audience before starting to translate reports or other support materials. 

• It may be helpful to get a sense of the language demand before going ahead with translation, e.g., by speaking to community contacts about the likelihood of community take up and whether the English version would suffice. If the demand is low, maybe the option to get an interpreter to attend a focus group session to discuss the content of the report would be another option.

• In some communities, members will only speak the language and will be not able to read. In situations like this getting an interpreter to read the report will also be a solution 

Optimise your files 

• Send the original editable source files, in this way you will avoid cost of formatting that could increase the cost of your translation.

• Some languages (e.g., Arabic, Mandarin, Cantonese, Hindi) will require different setting or formatting 

Interpreting project

• Which target languages do you want to prioritise? 

• When will the sessions take place? 

• How many participants do you intend to have? 

• Do you want to have one-to-one interviews or a group discussion? 

• When you are using interpreters, you need to think that you will need at least double the time of an interview without an interpreter. You will need to allow time for the interpreter to do his/he job. 

• It will be a good practice to create a poll to establish the number of participants which will attend the event and the languages required

• The location also could have an influence on the budget. If the session will be delivered out to the city, travel costs may occur. The agency will try to get the closest interpreters to attend the session. So, maybe it will be better to ask for a quote based on the location of the appointment. The agency will advise of the options available. 

Budgeting 

• Identify the financial resources available for your project. 

• Get a quote or some idea about the cost involved and make sure that you included the money for translation and interpreting support in your funding requests. 

• Establish a timeline, never leave translation to the last minute. This could increase the cost of the translation 

• Think about the location for the group discussion or interviews. If remote meetings using video are appropriate, this will cut the costs. 

Example costs for using In-trans at Grampian Regional Equality Council. Follow below our rates for interpreting and translations jobs. 

Translation

Cost of translation is based in the number of English words. Translation above 200 words: 

For languages Group 1 (Polish, Russian, Portuguese, Lithuania, etc) will be £0.10p per word plus VAT 

For languages Group 2 (Arabic, Hindi, Mandarin, Cantonese, etc) will be £0.12p per word plus VAT 

Translation under 200 words will be charged £30.00 

Formatting will be charged £30.00 per hour (if necessary) 

Proof of reading will be charged £0.08 per word (if necessary) 

Interpreting

The rate of interpreting is £30.00+ VAT for up to one hour during office hours (8:00-17:00) and after the first hour by the minute. 

There is no travel fee in the cases below: 

• in case we are able to provide an interpreter locally 

• travel need is less than 10 miles one way (20 miles in return) 

In other cases, travel expenses occur. We always inform you about it in advance, so that you can decide if you pay for the travel or prefer to change the date/ time of the appointment to have an interpreter locally. 

The charge of travelling time is £18.00 up to one hour and any time over one hour will be charged on a minute basis. The travel rate per mile is £0.48p 

Cancellation will be accepted at any time. However, by our policies, cancellations occurring 24 hours before the appointment will be charged for the first hour. An appointment cancelled over the 24 hours before the appointment will not be charged. 

Example costings 

These examples were provided in January 2022 by Dr Andrew Willis, Centre for Ethnic Health Research, University of Leicester, UK. 

Example costing 1. Animated recruitment video
Background
Production of a 2-3 minute recruitment animated video to support recruitment of participants with lower levels of English literacy to a large national trial taking place in an acute hospital setting. The animation and voiceover were co-designed and produced by patients and carers from ethnic minority communities during 3 workshops facilitated by designers and trial researchers. 

The team followed an established methodology for translation to maintain content validity, including clinician review to check clinical accuracy (Patel N, Willis A, Stone M, Barber S, Gray L, Davies M, et al. Developing a conceptually equivalent type 2 diabetes risk score for Indian Gujaratis in the UK. Journal of diabetes research. 2016;2016 ). The video was developed by an in-house design team using Vyond software (https://www.vyond.com). Translation of the script was done in-house, with two translators for each language, one of whom recorded the voiceover 

Co design workshops x 3, Video storyboarding, production and refinement, review of final version: £2500
Production of 2-3 min animated video in English, 2 stage translation dubbed into 4 languages (Polish, Urdu, Bengali and Punjabi) with pilot testing: £3000
Training sessions to cover video dissemination within the main trial: £750 

Total cost £6250 

Example costing 2. Questionnaire translation
Background
Translation of a (Approximately 2000 words) paper based self-assessment risk score for type 2 diabetes from English into Urdu. The research team followed an established translational model of firstly simplifying the English version and development of conceptual guidance for translators to follow. This was followed by forwards translation from English to Urdu using two qualified translators and then back translation from English to Urdu. Translators had to hold a diploma in public service interpreting and they were recruited from different translation companies locally. Once all translators were happy with the final version, it was reviewed by two Urdu literate clinicians and piloted with a group of 10-15 community members to check reading age, level of understanding and ease of use 

Researcher salary costs (recruitment, project management: £1750
Payments for translators: £2250
Community member/clinician vouchers: £250
Design costs: £750 

Total cost £5000 

The STRIDE project involved discussion in late 2022/early 2023 with ten panels of public contributors, trialists, health professionals and others. These panels discussed a total of 23 trials with the aim of saying which ethnic groups the trials needed to include and what proportion of the trial population they should be.

While considering a particular trial, panels sometimes provided advice on how a trial team might have answered the question about the ethnic groups to be included.  Additionally, having listened to ideas and suggestions from many individuals and panels across multiple trials and disease areas, the STRIDE team itself reflected on what might be useful advice for trial teams grappling with how to decide which ethnic groups their trial needs. 

This led to eight recommendations. Click here to read these recommendation and for more information. 

NIHR-INCLUDE was commissioned by the NIHR CRN to address the lack of representation in health and care research with the following vision:

‘provide better health and care through more inclusive research’

The NIHR-INCLUDE Guidelines provide:

• a strategic overview of potential points for intervention to improve inclusion of under-served groups across the life course of research; 

• key objectives to improve inclusivity; 

• guiding principles for stakeholders to apply the guidance. 

Some areas have been addressed at the level of individual communities or projects, whereas other areas require action at national or supra-national level, to provide appropriate regulatory, funding, governance guidance and support.

The NIHR INCLUDE frameworks can be found here.

This resource has been developed to support researchers to better understand how to embed EDI in research design and to meet the NIHR’s EDI requirements. The toolkit advocates for best practice which goes beyond the NIHR's current requirements. 

Click here to visit the the EDI Toolkit page.